New Magnetically Recyclable Reduced Graphene Oxide rGO/MFe2O4 (M= Ca, Mg)/Ag3PO4 Nanocomposites With Remarkably Enhanced Visible-light Photocatalytic Activity and Stability

Article abstract of DOI:10.1111/php.12968

New magnetically separable CaFe₂O₄/Ag₃PO₄, MgFe₂O₄/Ag₃PO₄, rGO/CaFe₂O₄/Ag₃PO₄ and rGO/MgFe₂O₄/Ag₃PO₄ photocatalysts were synthesized by the hydrothermal and ion-exchange deposition method. These four types of photocatalyst were used for degradation of Methylene blue (MB), Methyl orange (MO) and 4-chlorophenol (4-CP) in aqueous solution under visible-light illumination. The optimized photocatalyst, i.e. rGO/CaFe₂O₄/Ag₃PO₄ with a mass ratio of (1:3:9) composite not only shows the highest photocatalytic performance for the degradation of MB, MO and 4-CP under visible light irradiation among the other synthesized photocatalysts but also exhibits high reusability and stability for at least five cycles. It was found that the impressive separation of electron–hole pairs as well as presence of rGO sheets which act as a high speed charge transfer were responsible for increasing photocatalytic activity over the optimized photocatalyst under visible-light irradiation. A possible mechanism for the increased photocatalytic activity of the rGO/CaFe₂O₄/Ag₃PO₄ with a mass ratio of (1:3:9) composite was discussed in detail.

Full text of DOI:10.1111/php.12968

PROF. MOJTABA BAGHERZADEH (Orcid ID : 0000-0003-3009-3693)
Article type : Research Article
New Magnetically Recyclable Reduced Graphene Oxide
rGO/MFe O (M= Ca, Mg)/Ag PO Nanocomposites With
2
4
3
4
Remarkably Enhanced Visible-light Photocatalytic Activity and
Stability
Mojtaba Bagherzadeh*, Reyhaneh Kaveh
Department of Chemistry, Sharif University of Technology, PO Box 11155-3516, Tehran, Iran
Corresponding author’s email: bagherzadeh@sharif.edu (M. Bagherzadeh)
ABSTRACT
New magnetically separable CaFe O /Ag PO , MgFe O /Ag PO , rGO/CaFe O /Ag PO and
2
4
3
4
2
4
3
4
2
4
3
4
rGO/MgFe O /Ag PO photocatalysts synthesized by the hydrothermal and ion-exchange
2
4
3
4
deposition method. These four types of photocatalyst were used for degradation of Methylene
blue (MB), Methyl orange (MO) and 4-chlorophenol (4-CP) in aqueous solution under
visible-light illumination. The optimized photocatalyst, i.e. rGO/CaFe O /Ag PO with a
2
4
3
4
mass ratio of (1:3:9) composite not only shows the highest photocatalytic performance for the
degradation of MB, MO and 4-CP under visible light irradiation among the other synthesized
photocatalysts but also exhibits high reusability and stability for at least 5 cycles. It was
found that the impressive separation of electron-hole pairs as well as presence of rGO sheets
which act as a high speed charge transfer were responsible for increasing photocatalytic
This article has been accepted for publication and undergone full peer review but has not
been through the copyediting, typesetting, pagination and proofreading process, which may
lead to differences between this version and the Version of Record. Please cite this article as
doi: 10.1111/php.12968
This article is protected by copyright. All rights reserved.

activity over the optimized photocatalyst under visible-light irradiation. A possible
mechanism for the increased photocatalytic activity of the rGO/CaFe O /Ag PO with a mass
2
4
3
4
ratio of (1:3:9) composite was discussed in detail.
Keyword: Photocatalyst; Visible light; CaFe O ; MgFe O ; Ag PO ; rGO
2
4
2
4
3
4
INTRODUCTION
In the ubiquitous scientific world, development of environmentally friendly catalysts is
one of the more considerable issues. Semiconductor photocatalysis as a green technology to
remove hazardous organic contaminates in water has attracted extensive attention, as it can
use sunlight irradiation as the energy source [1–4].
In the past years, some new and efficient photocatalytic materials such as Ag-based
photocatalysts have been developed in order to increase the visible-light activity. For
instance, silver orthophosphate (Ag PO ) is an efficient visible-light responsive
3
4
photocatalyst, because of its perfect capabilities in the photodegradation of organic
contaminates. Ag PO has been considered a good semiconductor photocatalyst with high
3
4
photooxidative capabilities for O generation from water. Unfortunately, Ag PO is light-
2
3
4
sensitive and slightly soluble in aqueous media. Therefore, Ag PO can be photo-corroded
3
4
+
and decompose under light irradiation. Ag is transformed into weakly active Ag during the
photodegradation process, so photocatalytic activity of Ag PO slowly decreases due to the
3
4
photocorrsion by the photogenerated electrons and the practical application of Ag PO is
3
4
considerable limited [5–8].
In order to resolve this problem, recent studies have indicated that combining two or more
semiconductors to prepare a composite structure protects the Ag PO nanoparticles from
3
4
dissolution in aqueous media. This strategy can enhance photocatalytic performance and
stability of Ag PO [9–13].
3
4
Some Ag PO -based hybrid composites have been prepared in order to improve
3
4
photocatalytic
performance
and
stability.
For
example,
[15], Ag PO /Bi MoO
6
Ag/Ag PO4
[2],
[5],
3
TiO /Ag PO /graphene
[14], Fe O -Ag PO /AgI
2
3
4
3
4
3
4
3
4
2
Ag PO /nitridized Sr Nb O [16], Ag PO /Ag CO [17], Ag PO /CeO [18], Ag PO /rGO
3
4
2
2
7
3
4
2
3
3
4
2
3
4
[
19–21] have been synthesized and exhibited increased photocatalytic performance and
stability than neat Ag PO . However, aforementioned composites cannot separate and recycle
3
4
This article is protected by copyright. All rights reserved.

easily. To overcome this issue, magnetic composite photocatalysts have been developed.
These kinds of composites can separate from aqueous media via an external magnet easily
[
8,22].
Spinel ferrites have considered worthy of attention due to their magnetic properties and
the thermal and chemical stability of these particles. Spinel ferrites have been used for many
applications such as preparing efficient and separable composite photocatalysts [23–27].
Calcium ferrite (CaFe O ) and Magnesium ferrite (MgFe O ) are suitable choices for
2
4
2
4
fabrication of magnetic composite photocatalysts. CaFe O and MgFe O are visible-light
2
4
2
4
responsive semiconductors with narrow bands gap of approximately 1.9eV and 2.18eV,
respectively. Also, spinel ferrite-based catalysts can be easily separable because of their
magnetic properties [28–34].
Up to now, coupling of Ag PO with spinel ferrites is rare. Chen et al. fabricated a
3
4
ZnFe O /Ag PO photocatalyst which improved the photocatalytic activity compared with
2
4
3
4
neat Ag PO [35]. Zhao et al. synthesized NiFe O /Ag PO photocatalysts that enhanced the
3
4
2
4
3
4
photocatalytic performance towards the degradation of methylene blue [36].
In recent decades, reduced graphene oxide (rGO) has attracted tremendous attention in
photocatalysis because of its unique properties such as large surface area, better electronic
conductivity and good solution-dispersibility. In fact rGO is a good candidate for the
preparation of different graphene-based composite photocatalysts [37–41].
The extensive presence of organic contaminates in waste water from industries or
households is an serious environmental problem. These persistent organic chemicals can be
found in groundwater [8,19].
For instance, chlorophenols such as 4-chlorophenol (4-CP) in aqueous solution are
important micropollutants of the aquatic environment. Also, organic dyes (MB, MO, etc.) are
an important source of water contamination. Many technologies have been employed over the
years for water decontamination [42,43]. Among these various strategies, photocatalysis
phenomenon has attracted attention because of the simplicity, high reproducibility and perfect
performance.
This article is protected by copyright. All rights reserved.

The aim of this work is to prepare magnetically separable visible-light responsive
photocatalysts, which are stable and easy to synthesize. CaFe O /Ag PO , MgFe O /Ag PO ,
2
4
3
4
2
4
3
4
rGO/CaFe O /Ag PO and rGO/MgFe O /Ag PO composites were successfully synthesized.
2
4
3
4
2
4
3
4
The photocatalytic performance of these composites were evaluated by photo-degradation of
-CP, MB, MO in water under visible-light irradiation. The synergistic effects of GO
contents in CaFe O /Ag PO and MgFe O /Ag PO composites on the photocatalytic
4
2
4
3
4
2
4
3
4
efficiency are investigated.
The present study opens promising perspectives for increasing visible-light-driven
photocatalytic efficiency based on graphene oxide in order to destroy toxins.
MATERIALS AND METHODS
Reagents and apparatus. The materials including ferric nitrate (Fe(NO ) .9H O), calcium
3
3
2
nitrate (Ca(NO ) .6H O), magnesium nitrate (Mg(NO ) .6H O), silver nitrate (AgNO ),
3
2
2
3 2
2
3
sodium phosphate monobasic (NaH PO ), sodium hydroxide (NaOH) , graphite powder, MB
2
4
(
C H ClN S), MO (C H N NaO S), 4-chlorophenol (4-CP), were extra pure and
16 18 3 14 14 3 3
purchased from Merck company and used without further purification. Other chemicals and
solvents such as ethylene glycol (EG), ethanol (C H O) were of analytical grade and used as
2
6
received.
Fourier transform infrared (FTIR) spectra of the prepared samples were recorded on a
Bruker Tensor 27 spectrometer using a KBr pellet for sample preparation at room
temperature. The structure of samples were investigated via using X-ray diffraction (XRD)
technique that were carried out on an Inel Equinox 3000 X-ray diffractometer using Cu Kα
radiation. Raman measurements (wavelength 532 nm) were acquired by an Upright Raman
microscope with DPSSVD: YAG type of laser excitation. The magnetization of the prepared
samples was measured with vibrating sample magnetometer (VSM) (Lake Shore Model
7
400, Japan) operating at room temperatures. Optical absorption spectra of the prepared
samples were investigated by Ultraviolet−visible (UV−vis) diffuse-reflectance spectra (DRS)
on an Avaspec-2048-TEC UV−vis−NIR spectrometer that Barium sulfate (BaSO ) powder
4
was the internal standard in the wavelength range of 200−800 nm. All of the binding energies
were determined by this method. Photoluminescence (PL) spectroscopy was monitored using
a CARY ECLIPSE fluorescence spectrometer. The morphology and particle distribution were
This article is protected by copyright. All rights reserved.

performed by TESCAN MIRA ΙΙ field emission scanning electron microscope. The
electrochemical impedance spectra (EIS) were carried out by the Autolab
potentiostat/galvanostat apparatus in a three electrodes cell. Pt wire as the counter electrode,
Ag/AgCl (saturated KCl) as the reference electrode and 0.5 M Na SO solution as the
2
4
electrolyte were applied. EIS measurements were also gained at the open-circuit voltage and
these measurements were taken in the frequency range from 0.1 Hz to 100 kHz with AC
amplitude of 5 mV. In order to fabricate of working electrode, 30 mg as-prepared catalyst
powder and 1 mL ethanol solution were sonicated and then the mixture was sprayed on a 1×1
2
cm area of FTO (fluorine doped tin oxide) glass and allowed to dry at 400 °C for 1 h in
furnace. The EIS measurements were carried out at room temperature (25 °C). A 500 W
xenon lamp source equipped with AM 1.5G filter was used as the light source.
Synthesis of MFe O (M= Ca, Mg)/Ag PO photocatalysts. MFe O (M= Ca, Mg)/Ag PO
4
2
4
3
4
2
4
3
composites were synthesized by an ion-exchange method. In the first step, MFe O
2
4
nanoparticles were prepared by simple hydrothermal method. The calcium nitrate
Ca(NO ) .6H O and ferric nitrate Fe(NO ) .9H O with a molar ratio of (1:2) were added into
3
2
2
3 3
2
6
0 mL ethylene glycol (EG). After being sonicated for 10 min, the pH of the solution was
adjusted to 10 using an appropriate NaOH solution. Then the resulting mixture was
°
transferred into a 100 mL Teflon-lined autoclave and heated to 125 C for 24 h. After
filtration and being washed several times with ethanol, the product was dried to obtain
CaFe O nanoparticles. The MgFe O nanoparticles were synthesized by the same method by
2
4
2
4
replacing the calcium nitrate with magnesium nitrate.
A procedure for preparing of CaFe O /Ag PO composite with a mass ratio of (1:3) is as
2
4
3
4
follows:
.510 g (3 mmol) of AgNO was solved in 20 mL distilled water and ammonia aqueous
0
3
solution (1.0 M) was added gradually to the aforementioned solution in order to form a
transparent solution. Then 0.17 g CaFe O was added into the solution and resulting mixture
2
4
was sonicated for 30 min. 10 mL aqueous solution of NaH PO (0.1 M) was added with drop
2
4
by drop to the above mixture while sonication, then prepared mixture was sonicated for
another 30 min. The precipitates were filtered and being washed with distilled water and
dried at 80 °C for 24 h (Scheme 1). The MgFe O /Ag PO composite (1:3) were prepared via
2
4
3
4
the same procedure by replacing synthesized MgFe O nanoparticles with CaFe O
4
2
4
2
nanoparticles.
This article is protected by copyright. All rights reserved.

The Ag PO nanoparticles was prepared by a similar method without adding CaFe O or
3
4
2
4
MgFe O nanoparticles.
2
4
Synthesis of the rGO/ MFe O (M= Ca, Mg)/ Ag PO photocatalysts. Graphene oxide (GO)
2
4
3
4
was synthesized from purified graphite powder based on Hummers method [39]. In order to
synthesize rGO/ CaFe O composite a hydrothermal method was used. First Ca(NO ) .6H O
2
4
3 2
2
and Fe(NO ) .9H O with a molar ratio of (1:2) were dissolved into 60 mL EG and treated
3
3
2
with sonicate for 10 min. Second, 0.056 g GO was dispersed into 10 mL EG. After being
sonicated for 1 h, two solutions were mixed together and sonicated for another 1 h. Then
suitable amount of NaOH solution was added to resulting solution until the pH reached 10.
The mixture was transferred into a 100 mL sealed Teflon-lined autoclave for 24 h at 125 °C.
The prepared precipitate was washed with ethanol and dried to obtain rGO/ CaFe O
2
4
nanoparticles. Some amount of oxygen- containing groups in GO reduced during the
hydrothermal reaction process, so the products of rGO are in the prepared composite. The
similar method by replacing the calcium nitrate with magnesium nitrate was applied in order
to synthesize rGO/ MgFe O nanopaticles.
2
4
For typical preparation of rGO/MFe O (M= Ca, Mg)/Ag PO composites with a mass
2
4
3
4
ratio of (1:3:9), 0.510 g AgNO were dispersed into 20 mL distilled water and ammonia
3
aqueous (1.0 M) was added dropwise into the solution of AgNO . Then 0.17 g rGO/
3
+
MFe O (M= Ca, Mg) was added into the mixture sonicated about 2 h in order to make Ag
2
4
combine with the surface of rGO/ MFe O (M= Ca, Mg). After that, 10 mL NaH PO (0.1 M)
2
4
2
4
was added into the above solution gradually. The prepared mixture was sonicated for another
0 min. Finally, the product was collected and washed with distilled water and dried at 80 °c
3
for 2 h.
Photocatalytic activity measurement. The photodegradation performance of the synthesized
samples were evaluated for MB, MO and 4-CP degradation in aqueous solution under visible
light irradiation. A 250-W mercury lamp with a cut off filter (providing visible light 420
nm) as the light source was employed. Photocatalytic experiment was carried out in a 150 mL
photoreactor at room temperature (25 °C ± 2 °C). In all of the measurements, distance
between the lamp and the reactor was kept at 50 cm and its focused intensity on the reactor
-2
was calculated 85.272 µW.cm . Fig. 1 shows the emission wavelength of the 250-W mercury
lamp with the UV cut-off filter. For each experiment, a desired amount of prepared
photocatalyst (0.5-3 g/L) was added to the MB, MO and 4-CP aqueous solutions with a
This article is protected by copyright. All rights reserved.

favorite concentration of pollutant (15-35 mg/L) and illumination under continuous stirring.
Prior to irradiation, the solution was stirred in the dark for 1 h to obtain an equilibrium
between MB, MO and 4-CP aqueous solution and prepared photocatalysts. Then, the mixture
was exposed to Mercury lamp illumination under continuous stirring. At a given time interval
of irradiation, 5 mL of the reaction mixture was withdrawn and photocatalyst was
magnetically separated. The concentration of the residual MB, MO and 4-CP in the reaction
of solution were monitored by measuring the absorbance of solution at the wavelength 667,
4
69 and 280 nm respectively.
To study the reaction kinetic of photodegradation reaction of MB, MO and 4-CP under
visible light illumination, the pseudo-first-order model called Langmuir-Hinshelwood (L-H)
kinetic was used (Eq. (1)) [44].
(1)
−
1
−1
In which r, C, t, k and K are degradation rate (mg L min ), organic pollutant
−
1
concentration in aqueous solution after the various intervals of time (mg L ), irradiation time
−
1
−1
(
min), reaction rate constant (min ) and adsorption coefficient of organic pollutant (L mg ),
respectively. At low initial concentration of organic pollutant, the Eq. (1) is reduced to the
Eq. (2):
(2)
Where Kapp is gained for organic pollutant photodegradation from plotting the graph of
ln(C /C ) versus t.
0
t
The photocatalytic stability of the prepared photocatalysts was investigated by using
repeated test for MB, MO and 4-CP degradation. After reaction in each test, the mixture was
extracted to investigate the concentration of MB, MO and 4-CP, while the reacted prepared
photocatalysts were collected via an external magnet and dried at 80 °C for the next run.
Also, the free radical trapping experiments were carried out. Tert-butyl alchol (t-BuOH)
•
was chosen as the hydroxyl radical ( OH) scavenger, benzoquinone (BZQ) was chosen as the
•-
+
superoxide radical (O ) scavenger, KI was chosen as the hole (h ) scavenger.
2
This article is protected by copyright. All rights reserved.

RESULTS AND DISCUSSION
Characterizations
FT-IR spectroscopy. Fig. 2 displays FT-IR spectra of CaFe O , Ag PO , CaFe O /Ag PO
4
2
4
3
4
2
4
3
and rGO/CaFe O /Ag PO nanoparticles. As shown in Fig. 2a, the FT-IR spectrum exhibits
2
4
3
4
-1
the characteristic peaks of the calcium ferrite skeleton. The small peak appeared at 3408 cm
indicated the stretching vibration of the O-H band. Presence of this small peak is attributed to
-1
adsorbed water at the surface of calcium ferrite nanoparticles. Also, the band at 2501 cm is
-1
corresponding to the trace of atmospheric or adsorbed CO . The peak at 1433 cm can be
2
-1
attributed to the C-H bending frequencies of the remaining trace of EG. The band at 881cm
-1
correspond to Fe-O-H bending vibrations and the bands at 487 and 455 cm attributed to the
stretching vibration of the Ca-O and Fe-O (metal-oxygen bonds) [32,45].
-1
Fig. 2b shows the FT-IR spectrum of Ag PO nanoparticles. The band at 3176 cm is
3
4
-1
attributed to the O-H stretching of water molecules. The sharp peak at 1388cm is related to
3
-
the antisymmetric stretching mode of the PO4 group, whereas the symmetric vibration mode
3
-
-1
of PO4 group is found at 763cm . The asymmetric stretching vibration of P-O bonds is
-1
-1
found at 1020 cm . Also, the peak at 547cm is attributed to the asymmetric bending mode
of P-O bonds [46].
-1
In spectrum of CaFe O /Ag PO (1:3), the absorption peak at 422 and 462 cm ascribed
2
4
3
4
-1
metal-oxygen bonds that become weak and the absorption peaks at 555, 759 and 1020 cm
can be assigned to the symmetric bending mode of P-O bonds, symmetric vibration mode of
3
-
PO4 group and asymmetric stretching vibration of P-O bonds, respectively (Fig. 2c). In
-1
spectrum of rGO/CaFe O /Ag PO (1:3:9) composite (Fig. 2d), the peak at 1541 cm appear
2
4
3
4
due to the skeletal vibration of graphene [47].
Fig. S1a (see Supporting Information file) shows FT-IR spectra of MgFe O
2
4
-1
nanoparticles. The absorption bands at 3454 and 1629 cm are attributed to vibration of
-1
bonded O-H and H-O-H of adsorbed water hydroxyl ions. The band at 881 cm is related to
-1
Fe-O-H vibrations in FeOOH. The peaks at 490, 492 and 457 cm correspond to the
-1
vibrations of metal oxygen bonds such as Mg-O and Fe-O. The band at 1438 cm may be
related to C-H bending frequencies of remaining trace EG [45,48].
Fig. S1b exhibits the spectra of Ag PO nanoparticles and Fig. S1c displays the FT-IR
3
4
-1
spectra of MgFe O /Ag PO (1:3) composite. The peaks at 486 and 524 cm are assigned to
2
4
3
4
This article is protected by copyright. All rights reserved.

-1
the vibrations of metal-oxygen bonds. The bands at 557 and 767cm are attributed to the
3
-
symmetric bending mode of P-O bonds and symmetric vibration mode of PO , respectively.
4
-1
The peak at 1037 cm is related to asymmetric stretching vibration of P-O bonds. As shown
in Fig. S1d that is related to the sample of rGO/MgFe O /Ag PO (1:3:9) the band at 1541
2
4
3
4
-1
cm correspond to the skeletal vibration of graphene.
XRD analysis. Fig. 3 shows X-ray diffraction pattern of prepared Ag PO , CaFe O ,
3
4
2
4
CaFe O /Ag PO and rGO/CaFe O /Ag PO nanoparticles. All diffraction peaks can be
2
4
3
4
2
4
3
4
indexed to Ag PO (JCPDS NO. 6-0505) or CaFe O (JCPDS NO. 74-2136). The XRD
3
4
2
4
pattern of CaFe O /Ag PO (1:3) depicted in Fig. 3b indicates that the material obtained
2
4
3
4
contains just CaFe O and Ag PO . Also, at the XRD pattern of rGO/CaFe O /Ag PO
4
2
4
3
4
2
4
3
composite no typical peak of GO at 2θ~10 was observed that is due to the presence of
CaFe O and Ag PO nanoparticles between interlayers during the reduction process.
2
4
3
4
As shown in Fig. S2, the diffraction peaks of MgFe O are in agreement with standard
2
4
diffraction pattern of MgFe O (JPDC NO. 36-0398). The XRD pattern of MgFe O /Ag PO
4
2
4
2
4
3
(
1:3) indicated that this sample obtained contains MgFe O and Ag PO nanoparticles and no
2 4 3 4
other impurities are observed. In Fig. S2b, no peak of GO was detected. This issue is related
to the presence of MgFe O and Ag PO nanoparticles between interlayer of GO sheets
2
4
3
4
[
14,46,49,50].
FE-SEM images. Fig. 4a and Fig. S3a showed the FE-SEM micrograph of prepared CaFe O
2
4
and MgFe O nanoparticles, respectively. The micrographs showed the formation of spherical
2
4
CaFe O and MgFe O nanoparticles. CaFe O nanoparticles have diameter distribution
2
4
2
4
2
4
between 100-200 nm. The estimated particle size of MgFe O nanoparticles were in the range
2
4
of 50-70 nm. FE-SEM images of rGO/CaFe O and rGO/MgFe O composites were shown in
2
4
2
4
Fig. 4b and Fig. S3b, respectively. In a comparision, in Fig 4b, S3b homogenous spherical
particles of CaFe O and MgFe O were loaded on graphene’s surface. Presence of
2
4
2
4
graphene’s sheets could be vividly limited the aggregation and size of synthesized spinel
ferrites. It could be seen that prepared calcium ferrite had better interaction with graphene’s
sheets than magnesium ferrite.
Morphology of the prepared Ag PO nanoparticles was observed in Fig. 4c. As can be
3
4
seen from Fig. 4c, prepared Ag PO particles have irregular particles morphology.
3
4
This article is protected by copyright. All rights reserved.

The FE-SEM images of CaFe O /Ag PO and MgFe O /Ag PO composites were shown
2
4
3
4
2
4
3
4
in Fig. 4d and Fig. S3c, respectively. It can be seen that the CaFe O and MgFe O particles
2
4
2
4
scattered on the surface of Ag PO particles in aforementioned images.
3
4
Fig. 4e, f and Fig. S3d, S3e displayed the FE-SEM images of synthesized
rGO/CaFe O /Ag PO and rGO/MgFe O /Ag PO composites, respectively. The images of
2
4
3
4
2
4
3
4
the rGO/MFe O (M=Ca, Mg)/Ag PO with a mass ratio of (1:3:9) indicate that rGO sheets
2
4
3
4
are embedded between the crystal particles. The calcium and magnesium ferrite nanoparticles
are agglomerated around the Ag PO crystal to form a composite structure.
3
4
Raman analysis. Fig. 5 compares the Raman spectra of both CaFe O and rGO/CaFe O
4
2
4
2
nanoparticles. The CaFe O Raman plot shows the typical peaks at 440, 566, 649 and 1246
2
4
-1
cm , respectively. The Raman plot of rGO/CaFe O composites show the characteristics D
2
4
-1
-1
and G peaks at 1359 cm and 1571 cm in addition to the peaks that were related to CaFe O
2
4
nanoparticles.
Fig. S4 exhibites the Raman plot of MgFe O and rGO/MgFe O nanoparticles. The
2
4
2
4
-1
peaks in the range of 100-1000 cm were observed [43]. These peaks confirmed the pure
MgFe O nanoparticles. In Fig. S4b that is related to the rGO/MgFe O composites in
2
4
2
4
addition to the peaks corresponding to MgFe O , the characteristics D and G peaks at 1359
2
4
-1
-1
cm and 1563 cm appear which confirmed the dispersion of graphene oxide into reduced
graphene oxide.
VSM Magnetic analysis. The magnetic response of pure CaFe O , rGO/CaFe O /Ag PO and
2
4
2
4
3
4
MgFe O and rGO/MFe O /Ag PO (1:3:9) nanoparticles has been investigated by VSM.
2
4
2
4
3
4
(
Fig. 6 and S5). It could be observed that presence of MFe O
4
(M= Ca, Mg) nanoparticles
2
introduce the magnetic behavior in rGO/MFe O (M= Ca, Mg)/Ag PO composites. A drop in
2
4
3
4
M values has been observed from 30.2 emu/gm and 26.9 emu/gm for CaFe O and MgFe O
4
S
2
4
2
to 7 and 5 emu/gm for rGO/CaFe O /Ag PO and rGO/MgFe O /Ag PO nanoparticles,
2
4
3
4
2
4
3
4
respectively. The lower values of the M can be related to the presence of Ag PO particles
s
3
4
and rGO sheets in the rGO/MFe O (M= Ca, Mg)/Ag PO composites. However the
2
4
3
4
magnetic behavior of rGO/MFe O (M= Ca, Mg)/Ag PO composites allow their easy
2
4
3
4
separation from suspended system by an external magnet [51].
This article is protected by copyright. All rights reserved.

UV-vis analysis. Fig. 7a, b and Fig. S6a exhibits the UV-vis diffuse reflectance spectroscopy
of the CaFe O , Ag PO and MgFe O nanoparticles, respectively. The band gap energy (E )
2
4
3
4
2
4
g
of these three samples were calculated by the following equation:
1
/2
αhυ = A (hυ-Eg)
(3)
Where α is absorption coefficient and h is plank constant. Furthermore, υ and A are light
2
frequency and a constant, respectively. E can be obtain from a region in the plot of (αhυ)
g
versus photon energy (hυ) [37].
The band gap energy of the pure CaFe O , MgFe O and Ag PO nanoparticles are about
2
4
2
4
3
4
1
.52, 1.68 and 1.94 eV respectively. The E of these three samples are low compared to
g
standard value, which can attribute to the different conditions of synthetic hydrothermal
method. Diffuse reflectance spectroscopy of CaFe O /Ag PO (1:3) and MgFe O /Ag PO
4
2
4
3
4
2
4
3
(
1:3) composites are illustrated in Fig.7 and Fig. S6b, respectively. The value of E for
g
CaFe O /Ag PO composite was calculated 1.31 eV and this value for MgFe O /Ag PO
4
2
4
3
4
2
4
3
composite was 1.35 eV. So, these two composites have higher visible light absorption
capability. Introduce of rGO sheets decreases band gap energy of CaFe O /Ag PO and
2
4
3
4
MgFe O /Ag PO composites. The E for rGO/CaFe O /Ag PO and rGO/MgFe O /Ag PO
4
2
4
3
4
g
2
4
3
4
2
4
3
composites obtained 1.17 and 1.20 eV, respectively (Fig. 7d and Fig. S6c). Consequently,
rGO/CaFe O /Ag PO composite has lower band gap energy compare to other samples.
2
4
3
4
PL results. The Photoluminescence (PL) analysis has been widely performed to study the
migration, transfer and recombination processes of photo-induced electron-hole pairs in the
prepared samples. In fact, the lower the PL signal means that the higher separation efficiency
of photo-induced electron-hole pairs, implying that more photo-induced electrons and holes
can be available for redox processes. Fig. S7 displays room temperature PL spectra of the
pure Ag PO nanoparticles, as well as rGO/MFe O (M= Ca, Mg)/Ag PO (1:3:9) composites
3
4
2
4
3
4
upon the excitation wavelength at 420 nm. The main emission peak of Ag PO was vividly
3
4
decreased, owing to this fact that the recombination of photo-induced electron-hole pairs is
efficiency prevented because of the presence of MFe O nanoparticles and rGO sheets. As
2
4
shown in Fig. S7 the weak pL intensity of rGO/MFe O (M= Ca, Mg)/Ag PO (1:3:9)
2
4
3
4
composites, implying that the cooperative effects between MFe O /Ag PO nanoparticles and
2
4
3
4
rGO sheets contributed to enhancing the charge separation efficiency and decreasing the
recombination of photogenerated electron-hole pairs effectively [52].
This article is protected by copyright. All rights reserved.

Electrochemical impedance measurement (EIS). EIS analysis was employed to obtain further
insight into the charge transfer mechanisms. Fig. 8 exhibits the Nyquist plots of the
synthesized Ag PO and of rGO/MFe O (M= Ca, Mg)/Ag PO (1:3:9) composites at the
3
4
2
4
3
4
open-circuit voltage. The semicircle diameter of EIS is equal to total charge transfer
resistance (R ) in the prepared samples. As shown in Fig. 8 rGO/MFe O (M= Ca,
ct
2
4
Mg)/Ag PO (1:3:9) composites have smaller R than pure Ag PO nanoparticles, providing
3
4
ct
3
4
evidence that the rGO/MFe O (M= Ca, Mg)/Ag PO (1:3:9) composites have the low charge
2
4
3
4
transfer resistance and high efficient separation of photoinduced electron-hole [53].
Photocatalytic performance for MB, MO and 4-CP degradation
Evaluation of the photocatalytic activity. To evaluate the photocatalytic performance of the
synthesized samples, MB and MO photodegradation as well as 4-CP photodegradation were
carried out under visible irradiation. MB, MO and 4-CP are representative organic
contaminations. The control studies were also conducted without photocatalyst or light. The
photocatalytic degradation was calculated as follow:
%
Degradation = [(C -C )/C ] 100
(4)
0
t
0 *
Where C and C are concentration of organic contaminations before irradiation and after
0
t
irradiation, respectively.
First, the degradation efficiency for MB removal in the presence of prepared samples
under visible light illumination were investigated.
The amount of MB adsorbed onto CaFe O , CaFe O /Ag PO with a mass ratio of (1:3),
2
4
2
4
3
4
rGO/CaFe O /Ag PO (1:3:9) and MgFe O , MgFe O /Ag PO (1:3), rGO/MgFe O /Ag PO
4
2
4
3
4
2
4
2
4
3
4
2
4
3
(1:3:9) was approximately evaluated to be ~ 4%, 9%, 12% and 5%, 8%, 9% of the initial
amount of MB, respectively. As a result, the MB degradation without light irradiation could
be neglected. As shown, the absence of any photocatalyst also concluded in no degradation,
showing the removal of MB occurred through photocatalysis. Under aforementioned
conditions, no electron-hole pairs can be formed and therefore photocatalytic behavior of
synthesized samples is not induced. It is clear that the photocatalytic performances relate to
the formation, transfer and separation of photogenerated electron-hole pairs [53-55]. As
shown in Fig. 9, pure CaFe O degraded only 21% of MB after 45 min while,
2
4
This article is protected by copyright. All rights reserved.

CaFe O /Ag PO with a mass ratio of (1:3) showed 95% removal after 45 min that is related
2
4
3
4
to the combine CaFe O and Ag PO nanoparticles together. The CB potential of CaFe O is
2
4
3
4
2
4
more negative than that Ag PO , so the photogenerated electrons in CaFe O could migrate to
3
4
2
4
the surface of Ag PO and simultaneously, the photogenerated holes in the VB of Ag PO
4
3
4
3
could migrate to CaFe O . Aforementioned process could improve the separation of
2
4
photogenerated electron-hole pairs. Based on the degradation efficiency results, the
rGO/CaFe O /Ag PO (1:3:9) degraded 92% of MB within 27 min. The high photocatalytic
2
4
3
4
performance of rGO/CaFe O /Ag PO could be attributed to the efficient separation of
2
4
3
4
photogenerated electron-hole pairs. The rGO is a perfect electron acceptor and electron
transporter which increases the charge separation efficiency that avoid recombining the
photogenerated electron-hole pairs. Also, rGO keeps nanoparticles dispersed hence retards
agglomeration. Consequently, photocatalytic activity could improve. Pure MgFe O and
2
4
MgFe O /Ag PO with a mass ratio of (1:3) degraded 18% and 89% of MB after 45 min. The
2
4
3
4
rGO/MgFe O /Ag PO (1:3:9) composite showed 85% removal after 27 min (Fig. S8).
2
4
3
4
Photocatalytic degradation efficiency of MO in the presence of pure CaFe O ,
2
4
CaFe O /Ag PO (1:3), rGO/CaFe O /Ag PO (1:3:9) and pure MgFe O MgFe O /Ag PO
4
2
4
3
4
2
4
3
4
2
4,
2
4
3
(
1:3), rGO/MgFe O /Ag PO (1:3:9) under visible light irradiation are presented in Fig. 10
2 4 3 4
and Fig. S9, respectively. The amount of MO adsorbed onto CaFe O , CaFe O /Ag PO
4
2
4
2
4
3
(
1:3), rGO/CaFe O /Ag PO
(1:3:9) and MgFe O
MgFe O /Ag PO
(1:3),
2
4
3
4
2
4,
2
4
3
4
rGO/MgFe O /Ag PO (1:3:9) was evaluated to be ~ 6%, 5%, 9% and 5%, 6%, 8% of the
2
4
3
4
initial amount of MO, respectively. Consequently, the MO degradation without light could be
neglected.
It is cleared that with decreasing band gap energies of the synthesized composites,
photoactivity of prepared samples increase. In addition to MB and MO, 4-CP was selected as
another model organic pollutant in order to further investigated the photocatalytic activity of
the rGO/CaFe O /Ag PO (1:3:9) and rGO/MgFe O /Ag PO (1:3:9) that showed better
2
4
3
4
2
4
3
4
photocatalytic performance in dye degradation compare to CaFe O /Ag PO (1:3) and
2
4
3
4
MgFe O /Ag PO (1:3) composites. As shown in Fig. 11, after 160 min of irradiation, no
2
4
3
4
degradation of 4-CP observed without catalysts. The photocatalytic degradation of 4-CP in
the presence of rGO/CaFe O /Ag PO (1:3:9) and of rGO/MgFe O /Ag PO (1:3:9) under
2
4
3
4
2
4
3
4
visible light was 90% and 86% in about 160 min, respectively.
This article is protected by copyright. All rights reserved.

The photocatalytic kinetics of the synthesized photoctalysts were also studied. Kapp is
gained for photodegradation of MB, MO and 4-CP from plotting the graph of ln (C /C )
0
t
versus t as shown in Fig. 12, S10 and S11, respectively. Based on the k app values, one can
compare the photocatalytic performance of the various catalysts. The obtained results
displayed that the kinetic simulation curves are close to linear curves. All of the correlation
2
coefficients (R ) were higher than 0.9. Consequently, the pseudo first order kinetics model fit
the experimental data well [56,57]. As a result, the photocatalytic degradation of MB, MO
and 4-CP using synthesized photoctalysts were fitted using the pseudo-first-order L-H model.
As seen in Fig. 13, S12 and S13, the degradation rate of MB, MO and 4-CP in the presence of
rGO/CaFe O /Ag PO (1:3:9) under visible light illumination was higher than of others.
2
4
3
4
Effect of concentration of catalyst on pollutants photodegradation. The effect of
rGO/CaFe O /Ag PO (1:3:9) photocatalyst concentration on the photodegradation rate of
2
4
3
4
MB, MO and 4-CP were investigated in the range of 0.5 g/L to 3 g/L at constant pollutants
concentration (25 mg/L) and the obtained results are illustrated in Fig. 14, S14 and S15,
respectively. As shown, the photodegradation efficiency of MB, MO and 4-CP enhances
from 0.5 to 2 g/L. At low photocatalyst concentration, the adsorbed light controlled the
activity of photocatalyst. It could be attributed to the limited photocatalyst surface area. On
the other hand, further raising the photocatalyst concentration from 2 to 3 g/L has a negative
effect on the photodegradation efficiency. This issue may cause enhanced scattering and
turbidity effects, resulting in decreased visible light penetration on the synthesized
photocatalyst surface and into the solution. The results showed that the optimum
photocatalyst concentration for pollutants degradation was 2 g/L.
Effect of initial pollutants concentration on photodegradation. Fig. 14, S16 and S17 illustrate
the interaction effect of initial MB, MO and 4-CP concentration on the degradation rate of
pollutants by changing the initial concentration of pollutants from 15 to 35 mg/L with
constant photocatalyst concentration (2 g/L) under visible light irradiation, respectively. As
the pollutants concentration increases, the amount of pollutants adsorbed on the active
•
surface area of photocatalyst enhances. As a result, the formation of OH radicals will be
enhanced. The obtained results showed that the photodegradation efficiency of pollutants
decrease with increase in initial MB, MO and 4-CP concentration ranging from 25 to 35
mg/L. The increase in initial pollutants concentration decrease the visible light penetration to
the surface of photocatalyst. Furthermore, the limited active site onto the photocatalyst may
be another reason for the decreased photodegradation efficiency.
This article is protected by copyright. All rights reserved.

Effect of reactive species on photodegradation of pollutants. In order to further insight into
the photodegradation pathway of organic pollutants, the free radical trapping experiments to
investigate the role of reactive species in the photocatalytic system of
rGO/CaFe2O4/Ag3PO4 (1:3:9) were carried out. The degradation efficiency of MB, MO and
•
4
-CP under various scavengers (t-BuOH as the hydroxyl radical ( OH) scavenger, BZQ as the
•-
+
superoxide radical (O ) scavenger and KI as the hole (h ) scavenger) are shown in Fig.15,
2
S18 and S19, respectively [58].
As observed in Fig. 15, MB strongly degraded after 27 min photocatalytic treatment without
scavengers, while the addition of 1 mM t-BuOH scavenger caused fast deactivation of the
rGO/CaFe2O4/Ag3PO4 (1:3:9) photocatalyst, and the degradation efficiency of MB
decreased from 92% to 41% in 27 min. So, this result indicates that all ˙OH play main role in
this process. The introduction of 1 mM BZQ into the system also restrained the
photocatalytic activity of the photocatalyst toward MB after 27 min reaction, demonstrating
•-
that O play an important role in photodegradation. When the 1mM KI was introduced into
2
+
the system, the degradation rate was slightly reduced indicating that h has little effect on the
•
MB degradation. The same observations for MO and 4-CP degradation exhibit OH
•-
participation which was formed by oxidative and reductive reactions as well as O2 is the
great contributor for pollutants degradation.
Photocatalytic stability. Stability of the rGO/CaFe O /Ag PO with a mass ratio of (1:3:9)
2
4
3
4
that had higher photocatalytic activity as compared to other synthesized photocatalysts is
another essential factor for evaluating its performance. Therefore, rGO/CaFe O /Ag PO was
2
4
3
4
recycled for five times under the same conditions. The degradation of MB, MO and 4-CP
during each run is exhibited in Fig. 16a, b and c, respectively. The photocatalytic data
showed that photocatalytic activity of rGO/CaFe O /Ag PO (1:3:9) for organic
2
4
3
4
contaminations degradation changes little and even, 84%, 80% and 74% of MB, MO and 4-
CP is degraded after recycling for five times, respectively. In fact, after contact between
Ag PO and CaFe O particles, a p-n junction is generated, the charge carriers transfer from
3
4
2
4
CaFe O to Ag PO . Consequently, the rGO/CaFe O /Ag PO displays high stability because
2
4
3
4
2
4
3
4
it does not suffer from photocorrosion during the degradation contaminates. In addition, the
rGO sheets not only enhance the charge separation efficiency but also keep nanoparticles
dispersed hence retards agglomeration so the rGO/CaFe O /Ag PO photocatalyst shows high
2
4
3
4
This article is protected by copyright. All rights reserved.

stability. The used photocatalysts separated by an external magnetic field was washed with
distilled water for two times and dried at 80 °C for 1 h.
Possible photocatalytic mechanism in rGO/MFe O (M= Ca, Mg)/Ag PO system. In order to
2
4
3
4
elucidate photocatalytic degradation mechanism of organic pollutants over synthesized
•
•
+
photocatalysts, the reactive species including OH, O ‾ and h involved in the
2
photodegradation process were investigated. The photodegradation pathway of MB, MO and
4
-CP over the prepared photocatalysts could be inferred, which is illustrated in Scheme 2
-
+
[
59–62]. To clarify the separation of photo-induced e /h pairs over the synthesized
photocatalysts, it is essential to estimate the conduction band (CB) and valence band (VB)
potentials of the samples. These energy levels were calculated using the following equations:
e
ECB = χ - E – 0.5 E
(5)
(6)
g
EVB = ECB - Eg
e
Where E , E and E are the VB, CB potentials and the energy of free electrons vs.
VB
CB
hydrogen (4.5 eV), respectively. Also, χ is electronegativity of the semiconductor and it was
estimated via Eq. (13):
a
b
c 1/a+ b+ c
χ [χ (A) χ (B) χ (C) ]
(7)
=
In above equation a, b, and c are the number of atoms in the compounds [63]. The values of
E and χ for Ag PO are 1.94 and 5.98 eV, respectively. So, the E and EVB of Ag PO were
g
3
4
CB
3
4
estimated to be 0.51 and 2.45 eV versus normal hydrogen electrode (NHE). Moreover, the
estimated values of E and χ for CaFe O are 1.52 and 5.54 eV, respectively. As a result, E
CB
g
2
4
and EVB for CaFe O are 0.28 and 1.8 eV. E and χ for MgFe O are 1.64 and 5.67 eV,
2
4
g
2
4
respectively. Therefore, the ECB and EVB of MgFe O were estimated to be 0.35 and 1.99 eV.
2
4
For CaF O and MgFe O particles that are p-type semiconductors, the Fermi levels are close
2
4
2
4
to VB levels and for Ag PO particles which are n-type semiconductor, the Fermi level is
3
4
close to CB level.
Before contact between Ag PO and MFe O particles, the CB edge and Fermi level of
3
4
2
4
Ag PO are lower than those of MFe O . After contact when a p-n junction is generated, the
3
4
2
4
charge carriers transfer from MFe O to Ag PO . Consequently, an external electric field at
2
4
3
4
the heterojunction interface is generated [64]. The Fermi energy levels of n-Ag PO and p-
3
4
MFe O are aligned under equilibrium conditions. In fact, the photogenerated electrons in
2
4
This article is protected by copyright. All rights reserved.

MFe O could migrate to Ag PO and simultaneously, the photogenerated holes in the VB of
2
4
3
4
Ag PO could migrate to MFe O . Aforementioned process could improve the separation of
3
4
2
4
photogenerated electron-hole pairs. Consequently, the photogenerated electrons in CB of
•-
+
•
Ag PO react with surface adsorbed O to form O , which then reacts with H to form OOH,
3
4
2
2
•
followed by rapid decomposition to OH radicals. In addition, the photogenerated holes in VB
•
of MFe O could oxidize H O and OH‾ to form OH radicals, that are involved in the
2
4
2
•
photodegradation reaction of organic pollutants. In fact, the process of generating OH can
occur by two pathways.
rGO/MFe O /Ag PO shows significant advance in photodegradation of organic
2
4
3
4
pollutants compared to MFe O /Ag PO . The rGO sheets are perfect speed charge transfer
2
4
3
4
ability that enhances the charge separation efficiency which prevent recombination of the
photogenerated electrons and holes. The photogenerated electrons in MFe O migrate to the
2
4
rGO sheets due to marvelous electron conductivity and graphene sheets are good electron
acceptors. So, the electrons are quickly migrate to the graphene sheets. Then, the electrons
transferred to the surface of Ag PO (Fig. 17 and S20). As a result, the electron-hole
3
4
recombination of the rGO/MFe O /Ag PO is inhibited. In addition, rGO sheets can delay
2
4
3
4
aggregation of nanoparticles. So, the presence of rGO sheets in the composites can promote
the separation of electron-hole pairs that this issue has a main role in the increase quantity of
active holes for organic contaminations degradation also, the probability of combination of
photogenerated electrons and Ag PO decreases, that offers better recyclability and stability
3
4
of the prepared composites.
CONCLUSIONS
In summary, we have successfully indicated four efficient and magnetically separable
photocatalysts CaFe O /Ag PO , MgFe O /Ag PO , rGO/CaFe O /Ag PO and
rGO/MgFe O /Ag PO for MB, MO and 4-CP degradation. It was shown that
2
4
3
4
2
4
3
4
2
4
3
4
2
4
3
4
rGO/CaFe O /Ag PO with a mass ratio of (1:3:9) nanocomposite exhibits the wonderful
2
4
3
4
visible light photocatalytic activity compared with other synthesized nanocomposites, that
arises from synergistic effect between the prepared nanocomposites and rGO sheets.
On the basis of this study, the rGO/CaFe O /Ag PO nanocomposite photocatalytic
2
4
3
4
system is an effective and magnetically recoverable visible light driven photocatalyst for
degradation of MB, MO and 4-CP as representative of organic contaminations.
This article is protected by copyright. All rights reserved.

Acknowledgement
M.B. acknowledges the Research Council of Sharif University of Technology for the
research funding of this project.
SUPPORTING INFORMATION
Additional Supporting Information is available in the online version of this article:
Fig. S1. FT‐IR spectra of (a) MgFe O , (b) Ag PO , (c) MgFe O /Ag PO (1:3), (d)
2
4
3
4
2
4
3
4
rGO/MgFe O /Ag PO (1:3:9).
2
4
3
4
Fig. S2. XRD patterns of (a) MgFe O /Ag PO ; the iconograph in (a) is the XRD pattern of MgFe O
4
2
4
3
4
2
and (b) rGO/MgFe O /Ag PO .
2
4
3
4
Fig. S3. FESEM image of (a) neat MgFe O , (b) rGO/MgFe O , (c) MgFe O /Ag PO (1:3), (d, e)
2
4
2
4
2
4
3
4
rGO/MgFe O /Ag PO (1:3:9).
2
4
3
4
Fig. S4. Raman spectra of (a) MgFe O nanoparticles and (b) rGO/MgFe O nanocomposites.
2
4
2
4
Fig. S5. Magnetization curve of MgFe O , rGO/MgFe O /Ag PO (1:3:9).
2
4
2
4
3
4
Fig. S6. The UV–Vis DRS and the band gap for (a) MgFe O , (b) MgFe O /Ag PO (1:3), (c)
2
4
2
4
3
4
rGO/MgFe O /Ag PO (1:3:9).
2
4
3
4
Fig. S7. Photoluminescence spectra with an excitation of a diode laser (wavelength 420 nm) of
Ag PO , rGO/CaFe O /Ag PO (1:3:9), rGO/MgFe O /Ag PO (1:3:9).
3
4
2
4
3
4
2
4
3
4
Fig. S8. The degradation efficiency for MB (25 mg/L) removal in the process of the MgFe O ,
2
4
MgFe O /Ag PO and rGO/MgFe O /Ag PO nanocomposites under visible light irradiation
2
4
3
4
2
4
3
4
(
[catalyst]= 2 g/L).
Fig. S9. The degradation efficiency for MO (25 mg/L) removal in the process of the MgFe O ,
2
4
MgFe O /Ag PO and rGO/MgFe O /Ag PO nanocomposites under visible light irradiation
2
4
3
4
2
4
3
4
(
[catalyst]= 2 g/L).
Fig. S10. Linear-log plot for the photocatalytic degradation of MO in the presence of prepared
samples under visible light irradiation.
Fig. S11. Linear-log plot for the photocatalytic degradation of 4-CP in the presence of
rGO/CaFe O /Ag PO and rGO/MgFe O /Ag PO nanocomposites under visible light irradiation.
2
4
3
4
2
4
3
4
Fig. S12. The degradation rate constant of the fabricated catalysts for the photocatalytic degradation
of MO under visible light irradiation.
This article is protected by copyright. All rights reserved.

Fig. S13. The degradation rate constant of the fabricated catalysts for the photocatalytic degradation
of 4-CP under visible light irradiation.
Fig. S14. The degradation efficiency for MO (25 mg/L) removal using different
amount of rGO/CaFe O /Ag PO (1:3:9) under visible light irradiation.
2
4
3
4
Fig. S15. The degradation efficiency for 4-CP (25 mg/L) removal using different
amount of rGO/CaFe O /Ag PO (1:3:9) under visible light irradiation.
2
4
3
4
Fig. S16. The degradation efficiency for MO removal using different concentration of initial dye in
the presence of rGO/CaFe O /Ag PO (1:3:9) under visible light irradiation ([catalyst]= 2g/L).
2
4
3
4
Fig. S17. The degradation efficiency for 4-CP removal using different concentration of initial
pollutant in the presence of rGO/CaFe O /Ag PO (1:3:9) under visible light irradiation ([catalyst]=
2
4
3
4
2
g/L).
Fig. S18. Effect of different scavengers on degradation of MO by rGO/CaFe O /Ag PO under the
2
4
3
4
optimized conditions.
Fig. S19. Effect of different scavengers on degradation of 4-CP by rGO/CaFe O /Ag PO under the
2
4
3
4
optimized conditions.
Fig. S20. Schematic illustration of the proposed reaction mechanism for the photocatalytic
degradation of the dyes in the presence of rGO/MgFe O (M=Ca, Mg)/Ag PO nanocomposite.
2
4
3
4
References
1
. J. Fenoll, N. Vela, I. Garrido, G. Navarro, G. Lucas, S. Navarro (2015) Reclamation of
Water Polluted with Flubendiamide Residues by Photocatalytic Treatment with
Semiconductor Oxides. J. Photochem. Photobiol. 91, 1088–1094.
2
3
. X. Huang, J. Ding, Q. Zhong (2015) Catalytic decomposition of H O over Fe-based
2
2
catalysts for simultaneous removal of NO and SO . Appl. Surf. Sci. 326, 66–72.
X
2
. C. Guo, L. Song, Sh. Zhang (2018) Synthesis of Pb (PO ) /Ag PO Composite
9
4 6
3
4
Photocatalysts With Enhanced Visible Light Photocatalytic Activity. J. Photochem.
Photobiol. 94, 228-236.
4
. L. Zhang, R. Zheng, S. Li, B. Liu, D. Wang, L. Wang, T. Xie (2014) Enhanced
Photocatalytic H Generation on Cadmium Sulfide Nanorods with Cobalt Hydroxide
2
This article is protected by copyright. All rights reserved.

as Cocatalyst and Insights into Their Photogenerated Charge Transfer Properties. Appl.
Mater. Interfaces. 6, 13406−13412.
5
6
7
8
. Yang-S. Xu, Wei-D. Zhang (2013) Monodispersed Ag PO nanocrystals loaded on the
3
4
surface of spherical Bi2MoO6 with enhanced photocatalytic performance. Dalton
Trans. 42, 1094–101.
. Y. Bi, Sh. Ouyang, N. Umezawa, J. Cao, J. Ye (2011) Facet effect of single-crystalline
Ag PO sub-microcrystals on photocatalytic properties. J. Am. Chem. Soc. 133, 6490–
3
4
6
492.
. X. Zhao, L. Song, Sh. Zhang (2018) Synthesis of AgCl/Ag PO Composite Photocatalysts
3
4
and Study on Photodegradation Activity Based on a Continuous Reactor. J.
Photochem. Photobiol. https://doi.org/10.1111/php.12875.
. X. Chen, Y. Dai, J. Guo, T. Liu, X. Wang (2016) Novel Magnetically Separable Reduced
Graphene Oxide (RGO)/ZnFe O /Ag PO
4
Nanocomposites for Enhanced
2
4
3
Photocatalytic Performance toward 2,4-Dichlorophenol under Visible Light. Ind. Eng.
Chem. Res. 55, 568–578.
9
1
. X. Yang, H. Cui, Y. Li, J. Qin, R. H. Zhang, H. Tang (2013) Fabrication of Ag PO -
3
4
graphene composites with highly efficient and stable visible light photocatalytic
performance. Acs Catal. 3, 363–369.
0. B. Chai, J. Li, Q. Xu (2014) Reduced graphene oxide grafted Ag PO composites with
3
4
efficient photocatalytic activity under visible-light irradiation. Ind. Eng. Chem. Res.
3, 8744–8752.
5
1
1
1. Z. Yi, J. Ye, N. Kikugawa, T. Kako, S. Ouyang (2010) An orthophosphate semiconductor
with photooxidation properties under visible-light irradiation. Nat. Mater. 9, 559–564.
2. Jing-W. Xu, Zhi-D. Gao, K. Han, Y. Liu, Yan-Y. Song (2014) Synthesis of magnetically
separable Ag PO /TiO /Fe O
4
heterostructure with enhanced photocatalytic
3
4
2
3
performance under visible light for photoinactivation of bacteria. ACS Appl. Mater.
Interfaces. 6, 15122–15131.
1
3. W-S. Wang, H. Du, R-X. Wang, T. Wen, A-W. Xu (2013) Heterostructured Ag PO
3
4
/
AgBr/Ag plasmonic photocatalyst with enhanced photocatalytic activity and stability
This article is protected by copyright. All rights reserved.

under visible light. Nanoscale. 5, 3315–3321.
1
1
1
4. X. Yang, J. Qin, Y. Jiang, R. Li, Y. Li, H. Tang (2014) Bifunctional
TiO /Ag PO /graphene composites with superior visible light photocatalytic
2
3
4
performance and synergistic inactivation of bacteria. RSC Adv. 4, 18627–18636.
5. A.M. Idris (2014) An In-Situ Anion Exchange Method Synthesized of Ag PO
3
4
Functionalized with Fe O and AgI for Photocatalytic Degradation of Methyl Orange
3
4
under Visible Light Irradiation. Int. J. Mater. Sci. Appl. 3, 303–308.
6. J. Guo, H. Zhou, S. Ouyang, T. Kako, J. Ye (2014) An Ag PO /nitridized Sr Nb O
7
3
4
2
2
composite photocatalyst with adjustable band structures for efficient elimination of
gaseous organic pollutants under visible light irradiation. Nanoscale. 6, 7303–7311.
1
1
7. L. Ma, I. Jia, X. Guo, L. Xiang (2014) Current Status and perspective of rare earth
catalytic materials and catalysis. Chinese J. Catal. 35, 108–119.
8. R. Shao, X. Zeng, Zh. Cao, H. Dong, L. Wang, F. Wang, J. Liu, Zh. Li, Q. Liang (2015)
A novel Ag PO /Nb O fiber composite with enhanced photocatalytic performance
3
4
2
5
and stability. RSC Adv. 19, 673–679.
1
2
2
2
2
9. G. Chen, M. Sun, Q. Wei, Y. Zhang, B. Zhu, B. Du (2013) Ag PO /graphene-oxide
3
4
composite with remarkably enhanced visible-light-driven photocatalytic activity
toward dyes in water. J. Hazard. Mater. 244, 86–93.
0. B. Chai, J. Li, Q. Xu (2014) Reduced Graphene Oxide Grafted Ag PO Composites with
3
4
Efficient Photocatalytic Activity under Visible-Light Irradiation. Ind. Eng. Chem. Res.
3, 8744–8752.
5
1. Q. Xiang, D. Lang, T. Shen, F. Liu (2015) Graphene-modified nanosized Ag PO
3
4
photocatalysts for enhanced visible-light photocatalytic activity and stability. Appl.
Catal. B, Environ. 162, 196–203.
2. J. Zhao (2017) Environmental Energy-efficient fabrication of a novel multivalence
Mn O -MnO heterojunction for dye degradation under visible light irradiation.
3
4
2
Applied Catal. B, Environ. 202, 509–517.
3. E. Casbeer, V. K. Sharma, X. Z. Li (2012) Synthesis and photocatalytic activity of ferrites
under visible light: A review. Sep. Purif. Technol. 87, 1–14.
This article is protected by copyright. All rights reserved.

2
2
2
4. D. Guin, B. Baruwati, S.V. Manorama (2005) A simple chemical synthesis of
nanocrystalline AFe O (A= Fe, Ni, Zn): an efficient catalyst for selective oxidation of
2
4
styrene. J. Mol. Catal. A Chem. 242, 26–31.
5. Zh. Li, X. Lai, H. Wang, D. Mao, Ch. Xing, D. Wang (2009) General synthesis of
homogeneous hollow core-shell ferrite microspheres. J. Phys. Chem. C. Nanomater.
Interfaces. 113, 2792–2797.
6. M. Aamir, Muhammad N. Ashiq, Gh. Yasmeen, B. Ahmad, Muhammad F. Ehsan, T. He
(
2015) Synthesis and characterization of polyaniline/ZrCo-substituted nickel ferrite
NiFe Zr Co O ) nanocomposites: their application for the photodegradation of
(
1
.2 0.4
0.4
4
methylene blue. Desalin water treat. 57, 1–10.
2
7. H. Liu, H. Hao, J. Xing, J. Dong, Z. Zhang, Z. Zheng, K. Zhao (2016) Enhanced
photocatalytic capability of zinc ferrite nanotube arrays decorated with gold
nanoparticles for visible light-driven photodegradation of rhodamine B. J. Mater Sci.
5
1, 5872–5879.
2
2
8. William L. IV, J. Drwiega, David W. Mazyck, Seung W. Lee, W. Sigmund, Chang Y. Wu,
P. Chadik (2005) Magnetically agitated photocatalytic reactor for photocatalytic
oxidation of aqueous phase organic pollutants. Environ. Sci. Technol. 39, 8052–8056.
9. R. Dom, R. Subasri, K. Radha, P.H. Borse (2011) Synthesis of solar active
nanocrystalline ferrite, MFe O (M: Ca, Zn, Mg) photocatalyst by microwave
2
4
irradiation. Solid State Commun. 151, 470–473.
3
3
3
3
0. J. Cao, T. Kako, P. Li, S. Ouyang, J. Ye (2011) Fabrication of p-type CaFe O nanofilms
2
4
for photoelectrochemical hydrogen generation. Electrochem. commun, 13, 275–278.
1. Z. Zhang, W. Wang (2014) Solution combustion synthesis of CaFe O nanocrystal as a
2
4
magnetically separable photocatalyst. Mater. Lett. 133, 212–215.
2. L. Khanna, N.K. Verma (2013) Synthesis, characterization and in vitro cytotoxicity study
of calcium ferrite nanoparticles. Mater. Sci. Semicond. Process. 16, 1842–1848.
3. S. Verma, P.A. Joy, Y.B. Khollam, H. S. Potdar, S.B. Deshpande (2004) Synthesis of
nanosized MgFe O powders by microwave hydrothermal method Mater. Lett. 58,
2
4
1
092–1095.
This article is protected by copyright. All rights reserved.

3
4. T. Sasaki, S. Ohara, T. Naka, J. Vejpravova, V. Sechovsky, M. Umetsu, S. Takami, B.
Jeyadevan, T. Adschiri (2010) Continuous synthesis of fine MgFe O nanoparticles by
2
4
supercritical hydrothermal reaction. J. Supercrit. Fluids. 53, 92–94.
5. X. Chen, Y. Dai, W. Huang (2015) Novel Ag PO /ZnFe O composite photocatalyst with
3
3
3
4
2
4
enhanced visible light photocatalytic activity. Mater. Lett. 145, 125–128.
6. G. Zhao, L. Liu, J. Li, Q. Liu (2016) Efficient removal of dye MB: Through the
combined action of adsorption and photodegradation from NiFe O /Ag PO . J. Alloys
2
4
3
4
Compd. 664, 169-174.
3
3
7. B. Velaga, P. Shanbogh, D. Swain, Ch. Narayana, N. Sundaram (2018) High Surface
Area SnO -Ta O Composite for Visible Light Driven Photocatalytic Degradation of
2
2
5
an Organic Dye, J. Photochem. Photobiol. https://doi.org/10.1111/php.12896.
8. D. Das, A. Samal, J. Das, A. Dash, H. Gupta (2014) One-Pot Fabrication of RGO–
Ag VO Nanocomposites by in situ Photoreduction using Different Sacrificial
3
4
Agents: High Selectivity Toward Catechol Synthesis and Photodegradation Ability.
J. Photochem. Photobiol. 90, 57–65.
3
9. H. Liu, Y. Wang, L. Shi, R. Xu, L. Huang, Sh. Tan (2016) Utilization of reduced
graphene oxide for the enhancement of photocatalytic property of TiO nanotube.
2
Desalin water treat. 57, 13263–13272.
4
4
0. N. Zhang, Y. Zhang, Y. Xu (2012) Recent progress on graphene-based photocatalysts:
current status and future perspectives. Nanoscale. 4, 5792–5813.
1. D. Lang, T. Shen, Q. Xiang (2015) Roles of MoS and Graphene as Cocatalysts in the
2
Enhanced Visible-Light Photocatalytic H Production Activity of Multiarmed CdS
2
Nanorods. ChemCatChem. 7, 943 – 951.
4
2. J. Theurich, M. Lindner, D.W. Bahnemann (1996)Photocatalytic Degradation of 4-
Chlorophenol in Aerated Aqueous Titanium Dioxide Suspensions: A Kinetic and
Mechanistic Study. Langmuir. 12, 6368–6376.
4
4
3. W.S. Hummers Jr, R. E. Offeman (1958) Preparation of graphitic oxide. J. Am. Chem.
Soc. 80, 1339–1339.
4. E. Akbarzadeh, S. R. Setayesh and M. R. Gholami (2016) Synthesis of the visible-light-
This article is protected by copyright. All rights reserved.

driven Ag VO /Ag PO /Ag photocatalysts with enhanced photocatalytic activity. RSC
3
4
3
4
Adv. 18, 14909-14915.
4
4
5. Y. Shen, Y. Wu, X. Li, Q. Zhao,Y. Hou (2013) One-pot synthesis of MgFe O
2
4
nanospheres by solvothermal method. Mater. Lett. 96, 85–88.
6. G. Botelho, J. Andres, L. Gracia, L.S. Matos, E. Longo (2016) Photoluminescence and
Photocatalytic Properties of Ag PO Microcrystals: An Experimental and Theoretical
3
4
Investigation. Chempluschem. 81, 202–212.
4
7. N. Aina, W. Shaheena, B. Bashira, N. Abdelsalamb, M. Warsia, M. Khanc, M. Shahid
(
2016) Electrical, magnetic and photoelectrochemical activity of rGO/MgFe O
2 4
nanocomposites under visible light irradiation. Ceramics International. 42, 12401-
2408.
1
4
4
8. T.K. Pathak, N. H. Vasoya, V.K. Lakhani, K.B. Modi (2010) Structural and magnetic
phase evolution study on needle-shaped nanoparticles of magnesium ferrite. Ceram.
Int. 36, 275–281.
9. S. Zhang, J. Li, X. Wang, Y. Huang, M. Zeng and J. Xu (2014) In situ ion exchange
synthesis of strongly coupled Ag@ AgCl/g-C N porous nanosheets as plasmonic
3
4
photocatalyst for h ighly efficient visible-light photocatalysis. Appl. Mater. Interfaces.
, 22116-22125.
6
5
5
0. B. Jiang, Y. Wang, Jian-Q. Wang, Ch. Tian, W. Li, Q. Feng, Q. Pan, H. Fu (2013) In Situ
Fabrication of Ag/Ag PO /Graphene Triple Heterostructure Visible-Light
3
4
Photocatalyst through Graphene-Assisted Reduction Strategy. ChemCatChem. 5,
359–1367.
1
1. M.Z. Naik, A.V. Salker (2016) Tailoring the super-paramagnetic nature of MgFe O
2
4
3
+
nanoparticles by In incorporation. Mater. Sci. Eng B. 211, 37–44.
5
2. K. He, J. Xie, M. Li, X. Li (2018) In situ one-pot fabrication of g-C N nanosheets/NiS
3 4
cocatalyst
heterojunction with intimate interfaces for efficient visible light photocatalytic
H generation. Appl. Surf. Sci. 430, 208-217.
2
5
3. N. Mohaghegh, E. Rahimi (2016) BiPO photocatalyst employing synergistic action of
4
Ag/Ag PO nanostructure and graphene nanosheets. Solid. State. Sciences. 56, 10-15.
3
4
This article is protected by copyright. All rights reserved.

5
4. A. Kumar, A. Kumara, G. Sharma, M. Naushad, R. Cataluna Veses, A. Ghfar, F. Stadler,
M. Rizwan Khan (2017) Solar-driven photodegradation of 17-β-estradiol and
Ciprofloxacin from waste water and CO conversion using sustainable Coal-
2
char/polymeric-gC N /RGO metal free nano-hybrids. New J. Chem. 41, 10208-10224 .
3
4
5
5
5. H. Wang, L. Shen, F.Fu. Guo, Y. Liang (2016) Design and construction of the sandwich-
like Z-scheme multicomponent CdS/Ag/Bi MoO heterostructure with enhanced
2
6
photocatalytic performance in RhB photodegradation. New J. Chem. 40, 8614-8624.
6. X. Yuan, H. Wang, Y. Wu, G. Zeng, X. Chen, L. Leng, Zh. Wu, H. Li (2016) One-pot
selfassembly and photoreduction synthesis of silver nanoparticle-decorated reduced
graphene oxide/MIL-125(Ti) photocatalyst with improved visible light photocatalytic
activity. Appl. Organometal Chem. 30, 289–296.
5
7. Zh. Xue, T. Wang, B. Chen, T. Malkoske, Sh. Yu, Y. Tang (2015) Degradation of
Tetracycline with BiFeO Prepared by a Simple Hydrothermal Method. Materials. 8,
3
6
360-6378.
5
8. F. Zhang, Sh. Zou, T. Wang, Y. Shi, P. Liu (2017) CeO /Bi WO Heterostructured
2
2
6
Microsphere
with Excellent Visible-light-driven Photocatalytic Performance for
Degradation of Tetracycline Hydrochloride. J. Photochem. Photobiol. 93, 1154-1164.
5
9. A. Chithambararaj, N.S. Sanjini, A.C. Bose, S, Velmathi (2013) Flower-like hierarchical
h-MoO : new findings of efficient visible light driven nano photocatalyst for
3
methylene blue degradation. Catal. Sci. Technol. 3, 1405-1414.
6
0. Ch. Zhang, Y. Zhou, J. Bao, Y. Zhang, Sh. Zhao, J. Fang, W. Chen, X. Sheng (2017)
Hierarchical TiO nanosheet‐assembled nanotubes with dual electron sink functional
2
sites for efficient photocatalytic degradation of rhodamine B. Appl Organometal
Chem. 4204.
6
6
1. Q. Huang, M. Cao, Z. Ai, L. Zhang (2015) Reactive oxygen species dependent
degradation pathway of 4-chlorophenol with Fe@Fe O core-shell nanowires. Appl.
2
3
Catal. B Environ. 162, 319–326.
2. C. Yang, Q. Li, L. Tang, K. Xin, A. Bai, Y. Yu (2015) Synthesis, photocatalytic activity,
and photogenerated hydroxyl radicals of monodisperse colloidal ZnO nanospheres.
Appl. Surf. Sci. 357, 1928-1938.
This article is protected by copyright. All rights reserved.

6
6
3. N. Kaur, S. Kaur Shahi, V. Singh (2016) Choline Chloride Assisted Synthesis of N and
Metal Codoped TiO and their Photocatalytic Activity under Visible Light. J.
2
Photochem. Photobiol. 92, 69–75.
4. S. Sivakumar, A. Selvaraj, A. Kandasamy Ramasamy (2013) Photocatalytic Degradation
of Organic Reactive Dyes over MnTiO /TiO Heterojunction Composites Under UV-
3
2
Visible Irradiation. J. Photochem. Photobiol. 89, 1047–1056.
Figure Captions:
Fig. 1. Spectrum distribution of the Visible light source provided by a 250-W mercury lamp with a
UV cut off filter (visible light ≥420 nm).
Fig. 2. FT‐IR spectra of (a) CaFe O , (b) Ag PO , (c) CaFe O /Ag PO (1:3), (d)
2
4
3
4
2
4
3
4
rGO/CaFe O /Ag PO (1:3:9).
2
4
3
4
Fig. 3. XRD patterns of (a) Ag PO , (b) CaFe O /Ag PO ; the iconograph in (b) is the XRD pattern of
3
4
2
4
3
4
CaFe O and (c) rGO/CaFe O /Ag PO .
2
4
2
4
3
4
Fig. 4. FESEM image of (a) neat CaFe O , (b) rGO/CaFe O , (c) Ag PO , (d) CaFe O /Ag PO (1:3),
2
4
2
4
3
4
2
4
3
4
(
e, f) rGO/CaFe O /Ag PO (1:3:9).
2 4 3 4
Fig. 5. Raman spectra of (a) CaFe O nanoparticles and (b) rGO/CaFe O nanocomposites.
2
4
2
4
Fig. 6. Magnetization curve of CaFe O , rGO/CaFe O /Ag PO (1:3:9).
2
4
2
4
3
4
Fig. 7. The UV–Vis DRS and the band gap for (a) CaFe O , (b) Ag PO , (c) CaFe O /Ag PO (1:3),
2
4
3
4
2
4
3
4
(
d) rGO/CaFe O /Ag PO (1:3:9).
2 4 3 4
Fig. 8. EIS Nyquist plots of the Ag PO , rGO/CaFe O /Ag PO (1:3:9), rGO/MgFe O /Ag PO
4
3
4
2
4
3
4
2
4
3
(
1:3:9).
Fig. 9. The degradation efficiency for MB (25 mg/L) removal in the process of the CaFe O ,
2
4
CaFe O /Ag PO and rGO/CaFe O /Ag PO nanocomposites under visible light irradiation
2
4
3
4
2
4
3
4
(
[catalyst]= 2 g/L).
Fig. 10. The degradation efficiency for MO (25 mg/L) removal in the process of the CaFe O ,
2
4
CaFe O /Ag PO and rGO/CaFe O /Ag PO nanocomposites under visible light irradiation
2
4
3
4
2
4
3
4
(
[catalyst]= 2 g/L).
This article is protected by copyright. All rights reserved.

Fig. 11. The degradation efficiency for 4-CP (25 mg/L) removal in the process of the
rGO/CaFe O /Ag PO and rGO/MgFe O /Ag PO nanocomposites under visible light irradiation
2
4
3
4
2
4
3
4
(
[catalyst]= 2 g/L).
Fig. 12. Linear-log plot for the photocatalytic degradation of MB in the presence of prepared samples
under visible light irradiation.
Fig. 13. The degradation rate constant of the fabricated catalysts for the photocatalytic degradation of
MB under visible light irradiation.
Fig. 14. The degradation efficiency for MB removal using different amount of rGO/CaFe O /Ag PO
4
2
4
3
(
1:3:9) and different concentration of initial dye under visible light irradiation.
Fig. 15. Effect of different scavengers on degradation of MB by rGO/CaFe O /Ag PO under the
2
4
3
4
optimized conditions.
Fig. 16. (a) MB, (b) MO, (c) 4-CP degradation performance in consecutive cycles using the recycled
rGO/CaFe O /Ag PO (1:3:9) photocatalyst through magnetic separation.
2
4
3
4
Fig. 17. Schematic illustration of the proposed reaction mechanism for the photocatalytic degradation
of the dyes in the presence of rGO/CaFe O (M=Ca, Mg)/Ag PO nanocomposite.
2
4
3
4
This article is protected by copyright. All rights reserved.

This article is protected by copyright. All rights reserved.

This article is protected by copyright. All rights reserved.

This article is protected by copyright. All rights reserved.